We will deploy 4 new upper extremity neuroprostheses covering all elvels of cervical level spinal cord injuries (SCI) with sufficient numbers of subjects to provide support for larger clinical trials in the future. All of these neuroprostheses will use implanted stimulators and 3 of 4 2ill use implanted EMG recording modules. EMG-based control algorithms will provide fully implanted coman and control interfaces while exploiting the subjects'retained voluntary function. Both stimulation and recording odules will be commercially manufactured, greatly facilitating their eventual deployment in to the general clinical environment. This project will also push the frontiers of neuroprostheses in several significant ways. In collaboration with a parallel effort funded by NICHD, we will work to more "brain achine interfaces" into use as a neuroprosthesis command interface for individuals with high cervical spinal cord injusries. We will also examine the potential of useing the Flat Interface Nerve Interface (FINE) to activvate multiple muscles from as proximal a location as possible, potentially leading to a nerve electrode-only implementation with a lower surgical impact. Finally, we will examine the use of the FINE as an interface for recording from cutaneous enrve fascicles arising in the hand as a source of information for providing electrocutaneous feedback regarding the state of the hand.